The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: "Pieter Hoeben "
Subject: Re: Improving Rise/Fall times on Opto
Date: Mon, 06 Jan 2003 11:14:19 +0100
Organization: Hoeben Electronics
NNTP-Posting-Date: 6 Jan 2003 10:14:40 GMT
X-Newsreader: Forte Agent 1.92/32.572
What actually is done is keeping the voltage across the transistor
constant by a virtual (on a certain voltage level) shortcircuit to
prevent Miller effects on the capacitances of the base. The are
multiplied by the transistors gain and degrade the speed. As no
voltage changes, only current, no capacitors are (dis)charged...
On Fri, 6 Dec 2002 21:17:14 +1100, "davidjudy"
>"thorin" wrote in message
>> Dear Group, could anyone suggest an effective ciruit that will vastly
>> imrpove the rise and fall times on the output of a 4N35 or equv
>> curently I am using a 68K from base to emitter in the output transistor.
>> output transistor is connected as emiitter-follower to prevent the circuit
>> from inverting the input.
>> Is there a way of improving the sharpness of the pulse?
>> thanks in advance
>The secret is to use a circuit configuration the minimises the amount of
>voltage swing C-E. That translates into loading the opto with a very low
>If you have a -ve rail, connect the base of an NPN to GND, the emitter to
>the collector of the opto and the emitter of the opto to -ve supply
>(say -5V). The collector of the NPN goes to +ve supply via a load resistor.
>This is called a cascode connection. All the photo current of the opto wil
>lappear across the load on the NPN transistor, but Vce of the opto will
>remain virtually constant. A large value R (1M or so) across Vbe of the opto
>will also help speed it up. Make sure the transistor doesn't saturate. This
>scheme can be adapted to single rail by lifting the NPN base off GND with a
>resistive divider, at the expense of output swing.
>You can also use a grounded emitter NPN, with a turn on resistor base to +ve
>supply, and simply have the opto "short" the base to ground.
>Another arrangement is to use fast-ish bifet op-amp, like an LF356. With a
>single supply bias the + input to midrail with equal R's. Connect the opto
>between - input and gnd, and use a 100K pullup to +ve supply. Connect back
>to back 1N4148s from opamp o/p to - i/p. The diodes prevent the opamp from
>ever hitting the rails and saturating. You'll only get about 1v pp output
>swing, but a comparator stage after that will get you back to logic levels.
>BTW, a 4-pin opto such as PC811 is preferrable, because the base pin on a
>4N35 is very prone to noise pickup.
>Alternatively, buy a fast opto :-)
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup